University of Montana
ScholarWorks at University of Montana
Theses, Dissertations, Professional Papers
Graduate School
2016
Re-analyzing the function of demonstrative
reference in Tajik
Kelly E. Bowman
Follow this and additional works at: http://scholarworks.umt.edu/etd
Part of the Semantics and Pragmatics Commons
Recommended Citation
Bowman, Kelly E., "Re-analyzing the function of demonstrative reference in Tajik" (2016). Theses, Dissertations, Professional Papers.
Paper 10704.
This Thesis is brought to you for free and open access by the Graduate School at ScholarWorks at University of Montana. It has been accepted for
inclusion in Theses, Dissertations, Professional Papers by an authorized administrator of ScholarWorks at University of Montana. For more
information, please contact [email protected].
RE-ANALYZING THE FUNCTION OF DEMONSTRATIVE REFERENCE IN TAJIK
By
KELLY ELIZABETH BOWMAN
B.A. in Linguistics and Germanic Languages and Literature, University of Kansas
Lawrence, Kansas, 2013
Thesis
presented in partial fulfillment of the requirements
for the degree of
Master of Arts
in Linguistics
The University of Montana
Missoula, MT
May 2016
Approved by:
Scott Whittenburg, Dean of The Graduate School
Graduate School
Dr. Irene Appelbaum, Chair
Linguistics
Dr. Mizuki Miyashita
Linguistics
Dr. Rebecca Wood
Anthropology
Abstract
Bowman, Kelly, M.A., Spring 2016
Linguistics
Re-analyzing the function of demonstrative reference in Tajik
Chairperson: Dr. Irene Appelbaum
This thesis presents a re-analysis of Tajik demonstratives based on an alternative to the widely
accepted framework for understanding demonstrative reference. In this framework,
demonstrative reference is categorized according to two criteria: the anchor relative to which
reference is made, and the number of spatial distinctions the system has for encoding distance
from the anchor (Levinson 2004, O’Grady 2010). According to previous literature (Rastorgueva
1963, Perry 2005, Windfuhr & Perry 2009), Tajik has a speaker-anchored, two-way reference
system. However, these criteria alone do not account for the data presented in this thesis. I
therefore propose the following change to this categorization of demonstrative reference
systems: the spatial distinctions made by a system are not always based on distance from the
anchor. Instead they may be based on either distance or location. With a location-based
distinction, a referent is indicated based on its location relative to other potential referents, rather
than absolute distance from the anchor (Enfield 2003). Such a system is able to account for the
Tajik data. Data presented in this thesis also call attention to a restriction on the use of the distal
demonstrative not mentioned in existing descriptions of Tajik. The referent, if absent from the
discourse situation, must be part of the common ground, defined as the set of shared knowledge,
experiences, and beliefs common to discourse participants (Clark et al 1983). Finally, I propose a
re-analysis of the semantics of Tajik demonstrative reference in terms of whether multiple
demonstratives are present (Enfield 2003). In this view, Tajik has a basic demonstrative (glossed
as proximal) which simply indicates a referent, regardless of distance from the anchor. Meaning
distinctions, such as proximal/distal, are introduced only when multiple demonstratives are used,
in order to differentiate multiple potential referents.
ii
Acknowledgements
I would first like to acknowledge my thesis advisor, Dr. Irene Appelbaum, whose
extraordinary effort, encouragement, and dedication made the completion of this thesis possible.
Without her time, patience, and guidance I would not have been able to accomplish what I have.
I am extremely grateful to have had the opportunity to know and to work with her.
Next, I would like to thank the members of my committee: Dr. Mizuki Miyashita, who
has been especially influential to me during my time as a graduate student and has always
encouraged me to have confidence in myself and my work, and Dr. Rebecca Wood, whose
challenging questions were of great help in directing the revision process for this thesis.
I would also like to thank Dr. Leora Bar-el, whose course on Linguistic Field Methods
gave rise to this thesis. Having gone into her classroom with no fieldwork experience, it is only
with her exceptional instruction and thoughtful feedback that I was able to find the data
presented in this thesis.
In addition, this thesis would not have been possible without the help of Layokat
Rasulova, who was the language consultant during Dr. Bar-el’s field methods course. I would
especially like to thank her for her interesting and insightful comments, which often helped
direct the elicitation sessions, and for her patience with being repeatedly asked numerous very
similar sounding questions.
Along with all those already mentioned, I would like to thank Dr. Tully Thibeau, who
taught many of my courses at the University of Montana and who was always available to
answer questions whether or not they were directly related to coursework; my fellow graduate
students in the Linguistics Program, especially Dannii Yarbrough, for letting me bounce ideas
iii
back and forth with her, for telling me whether my sentences made sense, and for her genuine
and honest friendship; and the Linguistics Program’s undergraduate students, who displayed
inspirational enthusiasm and who always allowed me to believe that I was cool.
Finally, I would like to thank my parents, family, and friends. Without the influence,
care, and support of all of these people, I would not be the person I am today. I leave it to them
to decide whether or not that’s a good thing.
iv
List of Diagrams and Tables
DIAGRAM A................................................................................................................................................................21
TABLE1.........................................................................................................................................................................22
DIAGRAM B .................................................................................................................................................................23
TABLE2.........................................................................................................................................................................23
DIAGRAM C .................................................................................................................................................................25
TABLE3.........................................................................................................................................................................26
DIAGRAM D ................................................................................................................................................................27
TABLE4.........................................................................................................................................................................27
DIAGRAM E .................................................................................................................................................................30
TABLE5.........................................................................................................................................................................30
v
Table of Contents
Abstract ........................................................................................................................................... ii
List of Diagrams and Tables ........................................................................................................... v
Table of Contents ........................................................................................................................... vi
1. Introduction ................................................................................................................................. 1
2. Previous Research on Demonstratives in Tajik .......................................................................... 4
2.2
Rastorgueva (1963) ..........................................................................................................................5
2.3
Perry (2005) .....................................................................................................................................5
2.4
Windfuhr & Perry (2009) ................................................................................................................7
3. Approaches to the Study of Demonstratives ............................................................................... 9
3.1
General Approaches .......................................................................................................................10
3.2
A Location-based Approach (Enfield 2003) ..................................................................................14
3.3
Common Ground (Clark et al 1983) ..............................................................................................15
4. The Use of Demonstratives in Tajik ......................................................................................... 17
4.1
Diagrams A and B ..........................................................................................................................21
4.2
Diagram C ......................................................................................................................................25
4.3
Diagram D ......................................................................................................................................26
4.4
Diagram E ......................................................................................................................................29
5. Analysis of Tajik Demonstrative Usage ................................................................................... 34
5.1
Distance-Based vs Exophoric Usage .............................................................................................36
5.2
The Common Ground Constraint ..................................................................................................43
5.3
One Basic Demonstrative, Meaning Contrast, and the Dual Function of va͡ɪ ...............................44
6. Conclusions, Implications, and Issues for Further Research .................................................... 49
6.1
Categorizing Systems of Demonstrative Reference ......................................................................50
6.2
The Role of Common Ground in Demonstrative Reference .........................................................51
6.3
In as a Basic Demonstrative ...........................................................................................................53
6.4
Implications ...................................................................................................................................54
6.5
Limitations and Areas for Further Research ..................................................................................57
References ..................................................................................................................................... 64
vi
1. Introduction
The most basic use of demonstratives is to indicate a referent that is within the physical
context of a discourse, or the exophoric use of demonstratives (Diessel 2006, Levinson 2004,
Evans & Green 2006, Coventry et al 2014). Exophoric demonstrative reference is generally
described according to two criteria: (i) an anchor, to which spatial distinctions are relative; and
(2) the number of spatial distinctions a language makes. The view addressed in this thesis, which
appears to be widely shared (Levinson 2004, Diessel 2006, Hanks 2009, O’Grady 2010,
Coventry et al 2014), is that the spatial distinctions made with demonstrative reference are
universally based on distance from the anchor, and that the anchor is most often (though not
always) the speaker of an utterance. Additionally, at least one alternative claim about the nature
of the spatial distinction exists. Enfield (2003) proposes that, cross-linguistically, the spatial
distinction made in demonstrative reference is not based on the referent’s distance from the
anchor, but instead is based on the referent’s location relative to other potential referents in the
discourse situation. In this thesis I use language data from Tajik to show that the commonly
described distance-based interpretation of demonstrative reference is not cross-linguistically
applicable because it is unable to predict the function of demonstratives in the data. Similarly,
while Enfield’s (2003) location-based interpretation is able to account for the Tajik data, it is also
not universally applicable and does not account for the function of demonstrative reference in
every language. I propose that both interpretations of demonstrative reference, the distance-based
interpretation commonly described in the literature and the location-based interpretation
described by Enfield (2003) should be considered alternatives to each other; any given language
1
can be categorized as either distance-based or location-based. This additional consideration, I
argue, constitutes a third criterion for the classification of demonstrative reference systems.
Additionally, the data presented in this thesis show that, contrary to the existing
descriptions of demonstrative reference in Tajik, there is a non-spatial restriction on the use of
one of the Tajik demonstratives for referents that are absent from the discourse situation. In such
a case, the referent must be part of the common ground shared by the discourse participants.
Common ground refers to the collection of knowledge, information, beliefs, and experiences
mutually shared by the discourse participants. Restrictions on the use of demonstratives based on
common ground are not unique to Tajik. While common ground is not mentioned in many of the
general works on demonstrative reference in Tajik, it is described as a factor that affects
demonstrative reference in some other languages (Clark et al 1983).
The data presented in this thesis were collected as part of a semester-long course on
Linguistic Field Methods at the University of Montana in Spring 2015. The data were collected
during seven elicitation sessions over a period of approximately eight weeks. The data were
generated from a single consultant - an adult female native speaker of Tajik from Tajikistan, who
is also a fluent non-native speaker of English.
This thesis is organized as follows. In Chapter Two, I introduce the Tajik language and
present existing descriptions of Tajik grammar in English. In particular, I describe the
explanations of demonstrative reference in each source and show that those descriptions are
inadequate. In Chapter Three, I discuss existing notions of the use of demonstrative reference
cross-linguistically. I will present one very widely held view of how systems of demonstrative
reference work, as well as an alternative view which I argue better describes the data collected
for this thesis. In Chapter Four, I present the linguistic data on Tajik demonstratives I collected in
2
2015, which supports my central claim that the widely accepted interpretation of demonstrative
reference (described in Chapter Three) is not able to account for the function of Tajik
demonstratives. In Chapter Five I discuss in more detail the failure of the widely held
interpretation of demonstrative reference to account for the Tajik data, and show that an alternate
view of demonstrative reference (also presented in Chapter Three) better accounts for the Tajik
data. Finally, in Chapter Six, I use the analysis from the previous chapter to draw three main
conclusions that are supported by the Tajik data. Also in Chapter Six, I propose implications of
these conclusions within the field of Tajik language description as well as the field of linguistics
as a whole. I conclude Chapter Six with a discussion of the limitations faced while conducting
this research and ways in which further research could effectively build on the data collected for
this project.
3
2. Previous Research on Demonstratives in Tajik
2.1 Introduction
Tajik (also called Tajiki or Tajiki Persian) is an Indo-European language belonging to the
Persian branch of Southwestern Iranian. It is spoken mainly in Tajikistan and Uzbekistan by
approximately 7.8 million speakers (Ethnologue). It is sometimes described as a dialect of
Persian1. The canonical word order of Tajik is subject-object-verb. Since subjects are marked by
verb agreement and objects can be contextually implied, both subjects and objects may be
omitted (Windfuhr & Perry 2009). Tajik verbs have person, number, tense, aspect, and mood
inflections as affixes on the verb (Windfuhr & Perry 2009). According to Perry (2005), spoken
Tajik began to develop separately from the variety of Persian spoken in Iran in the sixteenth
century, while written Tajik remained unchanged until the early 1900s (Perry & Windfuhr 2009).
In 1928 the standard orthography of Tajik was switched to a system based on Latin characters
rather than Arabic, and this was again changed in 1939 when Cyrillic script became the standard.
This chapter provides a brief description of the three published English-language
grammatical descriptions of Tajik, which were consulted during the data elicitation process and
while writing this thesis. They are: Rastorgueva (1963), Perry (2005), and Windfuhr & Perry
(2009). Each of these Tajik grammars describes demonstrative reference, but none in much
detail.
All three of these sources provide the same general characterization of the system of
demonstrative reference in Tajik. Tajik demonstrative reference is described as making a two
1
In this work, I treat Tajik as an independent language.
4
way spatial distinction, encoded in the proximal demonstrative in, and two forms of the distal
demonstrative, on and va͡ɪ . It is also described as being a speaker-anchored system, in which
the choice of demonstrative is made based on the referent’s distance from the speaker.
Beyond these two generally agreed-upon features of Tajik demonstrative reference, the
three sources vary somewhat in their description of this phenomenon. Most of this variation has
to do with the level of descriptive detail found in each work. A summary of the description of
Tajik demonstrative reference according to each author is presented below.
2.2
Rastorgueva (1963)
Rastorgueva (1963) provides a very short and very basic description of demonstrative
reference in his sketch of Tajik grammar. According to Rastorgueva (and according to Perry
2005, see below), definiteness is expressed using demonstratives.
2.3
Perry (2005)
Perry (2005) briefly mentions “familiarity” or “known”-ness as a relevant feature similar
to what was described by the Tajik consultant in this thesis. This feature seems to be similar to
the concept of "common ground" as described by Clark et al (1983). This feature, according to
Perry (2005), plays a role in the interpretation of adjectival demonstratives, which are the focus
of this thesis. Specifically, Perry (2005) states that the referent of the proximal form in the phrase
in pisar ‘this boy’ is “’old information’ to both the speaker and the listener” (p. 67). This finding
is not reflected in the data I collected for this thesis. Moreover, the data presented in this thesis
suggest that it is in the determination of the referent of the distal form, va͡ɪ /that, not the referent
of the proximal form, in/this, that common ground plays an important role.
5
Demonstrative reference in Tajik as described by Perry (2005) is a speaker-anchored
system of reference which makes a two-way spatial distinction. The Tajik demonstratives
mentioned by Perry are in, glossed as the English demonstrative this and used for proximal
referents, and on or va͡ɪ , both glossed as the English that and both used for distal referents.
According to Perry (2005), va͡ɪ is used only in the spoken language, and its use can be
considered identical to the use of on. The two are said to be used interchangeably. Perry (2005)
also addresses the metaphorical use of Tajik demonstratives to refer to something at a specific
point in time, stating that the proximal demonstrative in ‘this’ is generally used to refer to
something in the present or future, and the distal demonstrative on ‘that’ is generally used to
refer to something in the past.
Finally, Perry states that proximal in/this tends to refer to something “specific and
actual”. He provides an example of this using the phrase, to this person, who has rendered you
great assistance, noting that this phrase is used as a referential expression, meaning that it
indicates a specific person. Similarly, he states that on/that tends to be used to refer to something
nonspecific or hypothetical, providing the example of one who/he who/whoever constructions,
which are attributive expressions, meaning that they are nonspecific in that they indicate their
referent based on a description rather than indicating a specific referent. The referent of such an
expression is whichever entity matches the description contained within the expression. My data,
by contrast, seem to show that va͡ɪ does not behave in the same way as on in this case.
According to my data, the referent of va͡ɪ must be an entity that is specific and that is within the
common ground of the speaker and listener.
6
2.4
Windfuhr & Perry (2009)
Much of the information included in Windfuhr & Perry’s (2009) description of Tajik is
identical to that presented in Perry’s (2005) description. However, this description as a source
also serves to bring attention to the classification of Tajik by some authors as a dialect of Persian.
Detailed descriptions of the system of demonstrative reference in Persian seem to be similarly
rare2. While this thesis does not make any attempt to argue for or against Tajik being a separate
language or a dialect of Persian3, the classification of Tajik by some authors as a dialect of
Persian could explain the rarity of background literature pertaining specifically to Tajik.
Tajik demonstrative reference as described by Windfuhr & Perry (2009) makes a twoway distinction that is speaker-anchored. This source also describes the in and on/va͡ɪ uses in
approximately the same way that it is described in Perry (2005), but in slightly less detail. They
also state that in and on are “weak demonstratives” and “function virtually as definite articles”
(Windfuhr & Perry 2009). While this characterization of the Tajik demonstratives seems to be
compatible with Rastorgueva’s (1963) description of the Tajik demonstratives as functioning
purely to indicate definiteness, in this thesis I argue against this characterization.
While there are a few grammatical descriptions of Tajik, they do not go into great detail
on the use of demonstrative reference. Moreover, a number of the claims made in these
grammars are in conflict with the findings presented in this thesis.
2
Descriptions of Persian and Tajik that are written in English are rare. This may or may not be the case for
descriptions written in other languages.
3
While I treat Tajik as an independent language, it is not the purpose of this thesis to make any argument on that
point.
7
3. Approaches to the Study of Demonstratives
In a broad sense, deixis refers to contextual aspects of a discourse situation that influence
how a speaker’s utterance is understood by the listener. According to Hanks (1992) this can
include “any aspect of utterance form whatsoever,” (p. 46) including not only spatial or temporal
details of the discourse situation, such as the time of utterance or the relative positions of the
discourse participants and any surrounding objects to each other, but also factors such as the
speaker’s accent or intonation. Deictic expressions pick out a referent based on its position
within the discourse context.
Demonstrative reference refers to a set of deictic expressions, including demonstratives
such as this and that, pronouns, and adverbs. These expressions refer to objects or entities by
describing them in terms of some relative aspect of the discourse context. This means that
expressions of demonstrative reference refer to an entity by describing its relative position within
the discourse situation. A referent can be described according to its relative physical position,
such as the physical location of the referent relative to the speaker or listener or some other
relevant entity. This is most often interpreted to be the referent’s distance from the speaker or
listener. Or a referent can be described according to its relative metaphorical position (Diessel
2006, Evans & Green 2006). This can include the referent’s temporal position relative to the time
of utterance or relative to the time of some other event relevant in the discourse. Demonstrative
use can also be influenced by the relative importance of the referent (according to the speaker)
within the discourse. (Leonard 1982, Kresin 1998)
8
3.1
General Approaches
Evans & Green (2006) describe the function of demonstrative reference by focusing on
the deictic center and its influence on a listener’s perception of an utterance, rather than on any
direct physical relationship between an utterance and its referent. The deictic center is the central
spatial or temporal point to which all deictic expressions are relative. It is often identical to the
location of the speaker at the time of utterance (Diessel 2006). However, the deictic center can
vary in size (from including only the space occupied by the speaker, to including the space of a
room, a single building, an entire city, a country, etc.) and can also be established by the speaker
to be some other point in space and/or time that does not include the speaker or the listener and is
independent of the immediate discourse context (Maienborn et al 2012). Additionally, in a
language in which the system of demonstrative reference considers the location of the listener as
well as the location of the speaker, the deictic center may be assumed by a speaker of that
language to include the listener, either with the speaker or without the speaker, in cases when the
use of a deictic expression is anchored to the listener.
According to Diessel (2006), demonstratives have two basic functions, which are related:
1) to make direct reference to some particular entity based on its physical location relative to the
deictic center, and 2) to direct the joint attention of all participants in a discourse. Diessel (2006)
and Coventry et al (2014) both describe the spatial use of demonstratives to refer to physical,
tangible entities by indicating their location in terms of distance from the deictic center (what
Diessel calls the “exophoric” use of demonstratives) as “prototypical” (Levinson 2004, as cited
in Diessel 2006) and “basic” (Coventry et al 2014). Diessel additionally notes that the exophoric
use of a demonstrative is often (and in some languages necessarily) accompanied by some sort of
indicative gesture, such as pointing to or gazing at the referent, which provides a visual cue to
9
direct the attentional focus of the listener. Both further describe metaphorical, non-spatial, and
non-physical uses of demonstrative reference as “extensions” of (Diessel 2006) or “derived”
from (Coventry et al 2014) the original, basic, exophoric use. Evans & Green (2006) also support
the idea that non-physical and non-spatial uses of demonstratives are extended or derived by
suggesting that such uses are simply metaphorically, rather than physically, “pointing to” some
entity or some portion of a prior utterance on which the speaker wishes the listener to focus their
attention.
Hanks (2009) describes demonstrative reference as consisting of three parts: an object
(the referent that is picked out), an origo (the entity relative to which a referent is picked out),
and the relation between the referent and the origo (which is often described as the physical
distance between the two). According to what Hanks (2009) calls the “egocentric” understanding
of spatial deixis, the origo or deictic center is the speaker and the relation is based on the
referent’s physical distance from the origo. The egocentric version is one way that
demonstratives can be used, but Hanks also describes demonstratives as “sensitive” to many
other factors, including common ground and ownership.
Pederson et al (1998) describe different frames of reference used for deixis, including
absolute reference, which is fixed based on contextual information about the discourse situation
independent of the discourse participants (such as cardinal directions); relative reference, based
on the referent’s position relative to some other point (the deictic center, which they state as most
often being egocentric); or intrinsic reference, including languages in which there are no separate
terms to describe spatial positioning (languages in which there may be terms for things such as
left hand and right hand in the same way that there are terms for any other body part, but in
which there are no separate terms for left or right). They categorize languages as being intrinsic,
10
absolute, relative, or mixed. Despite common arguments that the relative frame of reference with
a human body as its center, often the speaker, is linguistically universal (Clark 1973, Lyons
1977, and Miller & Johnson-Laird 1976, all as cited in Pederson et al 1998), the results of
Pederson et al’s (1998) cross-linguistic study of spatial reference suggest that this is not the case,
with more than half of the thirteen languages included in the study being categorized as mixed
(using both relative and absolute reference).
In language grammars, demonstratives are generally described as composing a system for
making spatial distinctions based on distance. Spatial demonstratives do not share exactly the
same characteristics or the same usage conventions across all languages. A specific language’s
system of demonstrative reference can be categorized based on two criteria, as described by
Levinson (2004) and O’Grady (2010): the number of spatial or distance distinctions made in the
language, and on what basis those distinctions are based or anchored. Languages commonly use
proximal/distal two-way distinctions or proximal/medial/distal three-way distinctions. Other
systems are also attested, such as the speaker-proximal/listener-proximal/speaker- and listenerdistal system used in Japanese. The anchor is typically the location of the speaker (speakeranchored), the location of the addressee (addressee-anchored), or the locations of both speaker
and addressee (speaker/addressee-anchored) (Levinson 2004). In particular, Levinson (2004)
specifies that while two-term systems are only speaker-anchored, systems in which there are
three or more terms can be speaker- or speaker/addressee-anchored. Below is an example of a
comparison between a system with a speaker-anchored, three-way distinction (Spanish) and a
system with a speaker/addressee-anchored, three-way distinction (Japanese):
11
(1)
Spanish
Distance from speaker
(proximal)
este
+
(medial)
ese
++
(distal)
aquel
(Levinson 2004)
(2)
Japanese
Speaker distance
+
+
(Levinson 2004, O’Grady 2010)
Addressee distance
+
+
kosoa-
Despite their usage differences, these sets of terms can be glossed the same way in English:
este/ko-/this, ese/so-/that, and aquel/a-/that over there.
The distinction between proximal and distal is based on whether or not the referent is
within the anchor’s proximal space or near-space. The boundary of this space is flexible in that it
can be defined by the speaker of an utterance or some specific contextual aspect of a discourse.
Generally, a referent within the anchor’s peripersonal space is considered proximal. The
peripersonal space of a discourse participant is the space within reach of the participant. The
space beyond that area is the participant’s extrapersonal space (Kemmerer 1999).
There are three types of spatial uses of demonstratives: adjectival (e.g., I want to cuddle
that puppy), pronominal (e.g., I can’t get over that), and adverbial (e.g., the puppy ran there)
(Kemmerer 1999). The examples used in this thesis consist of adjectival uses, although the
consultant noted that the answers provided during the elicitation regarding the use of the
adjectival demonstratives could also be applied to the pronominal use of the same
demonstratives. As an example, the consultant indicated that her answers regarding the use of the
proximal demonstrative va͡ɪ (glossed by Rastorgueva 1963, Perry 2005, and Windfuhr & Perry
12
2009 as English that) in the phrase va͡ɪ sag ‘that dog’ are also applicable to the use of va͡ɪ as a
third person pronoun (meaning it).
The problem with this approach to demonstrative reference is that it fails to predict the
results of the Tajik data elicitations presented in this thesis. While in many situations the Tajik
demonstratives function in ways that this interpretation of demonstrative reference would
predict, they also function in some unexpected ways. The speaker’s choice of demonstrative for a
referent in a distal position, between the speaker and the listener but outside the peripersonal
space of both, varies depending on other circumstances within the discourse situation. The use of
the demonstrative glossed as near or proximal under some circumstances and the far or distal
demonstrative under other circumstances for a referent that is the same distance away from the
speaker cannot be explained by a distance-based demonstrative reference system.
3.2 A Location-basis for Demonstrative Reference (Enfield 2003)
Every language has at least two demonstratives, including a proximal and a distal.
Enfield (2003) describes the basic function of demonstratives as indication, suggesting that in
each language there is one basic or unmarked demonstrative word and its function is simply that
it points to a referent. As Enfield (2003) states, differing arguments have been made in the
literature as to which demonstrative, the proximal or distal, is typically the basic or unmarked
one. It is specifically the inclusion of additional demonstratives that adds further meaning to a
system of reference. This is described by Enfield in terms of determiners in Lao, nii and nan,
which are respectively glossed as DEM (the gloss used by Enfield for the basic unmarked
demonstrative) and as DEM not here. According to this interpretation of demonstrative
reference, the basic demonstrative nii simply points to a referent, and the additional
13
demonstrative nan also points to a referent with the additional stipulation that the referent is
absent from the discourse situation.
In addition, Enfield (2003) describes an interpretation of the spatial distinction made by
demonstrative reference that differs from the one presented in section 3.1. According to Enfield
(2003), the spatial distinction of demonstrative reference is not based on the referent’s distance
from the anchor. Instead, it is based on the referent’s position in the discourse situation relative
to other potential referents. In this way, according to Enfield (2003), the use of a demonstrative
expression is not for the purpose of describing how far away a referent is, but instead its purpose
is to describe which entity the referent is. The difference between these two interpretations of
demonstrative reference will become the basis for part of my data analysis in Chapter Five.
3.3 Common Ground (Clark et al 1983)
Common ground refers to the body of knowledge and information that is shared by all
participants of a discourse. Information in the common ground comes from multiple sources,
including joint perceptual experience, jointly experienced hearsay, and membership within a
particular community or in-group (which includes everything that a participant believes to be
universally known or believed within that community or group). Common ground knowledge
does not have to be established before a discourse takes place; for example, two complete
strangers engaging in a conversation have common ground, though it may be very limited.
According to Clark et al (1983), “the speaker intends each addressee to base his inferences not
on just any knowledge or beliefs he may have, but only on their mutual knowledge or beliefs –
their common ground.” (p. 247) This means that when a speaker utters a demonstrative
expression, the listener’s identification of its referent is based upon which potential referent is
most salient based on the common ground they share with the speaker.
14
The importance of common ground in demonstrative reference will also play a role in my
data analysis. I will claim that the status of the intended referent of a demonstrative phrase (as
inside or outside of the speaker’s and listener’s common ground) plays an important role in the
use and interpretation of demonstratives in Tajik. In particular, the common ground status, I
argue, is important in the use and interpretation of the Tajik distal demonstrative va͡ɪ .
15
4. The Use of Demonstratives in Tajik
This chapter shows data elicited from a language consultant using a series of picture
diagrams. The data show ways in which demonstrative use in Tajik differs from demonstrative
use in English. Based on the categorization of Tajik demonstrative reference as a speakeranchored system with a two-way distance distinction, and based on the description of Tajik
demonstratives (Rastorgueva 1963, Perry 2005, Windfuhr & Perry 2009), the function of Tajik
demonstratives is unexpected.
There are three demonstratives in Tajik: in and on or va͡ɪ . These are described in a
number of grammar sketches of Tajik as proximal and distal demonstratives respectively
(Rastorgueva 1963, Perry 2005, Windfuhr & Perry 2009). The form in is described as the
proximal demonstrative. Both on and va͡ɪ are described as distal. According to Perry (2005), on
and va͡ɪ have the same meaning and are used interchangeably. The data collected in this thesis
concern in and va͡ɪ . My thesis does not include any data or discussion related to on. This is
because the data elicited from the consultant over the course of this project, when asked to
produce utterances in Tajik, included instances of va͡ɪ , but not of on. The Tajik demonstratives
in (proximal) and va͡ɪ (distal) correspond to the English spatial demonstratives this (proximal)
and that (distal) and are glossed as such, as shown in the examples below (Rastorgueva 1963).
(3) Tajik Demonstratives and English Gloss
(a)
in
xona
this
house
(b)
on
that
xona
house
va͡ɪ
xona
that
house
(Rastorgueva 1963)
(c)
16
According to the features of demonstrative reference described by Levinson (2004), the
systems of demonstrative reference in English and Tajik can both be classified in the same way.
Both systems are speaker-anchored, meaning that the speaker chooses the appropriate
demonstrative to use for a particular referent based on the referent’s physical location relative to
the speaker. Both systems also make a two-way spatial distinction, consisting of proximal and
distal terms. However, examples that I elicited show that, despite these formal similarities, the
Tajik demonstratives function differently than the demonstratives of English.
The data presented in this thesis were collected from a single language consultant, who is
a native speaker of Tajik and a fluent non-native speaker of English. Data were elicited using a
series of picture diagrams. Each of the diagrams presented to the consultant depicted a discourse
situation involving a speaker, a listener, and up to three dogs. In each diagram, the dogs were
positioned at various distances from the speaker and from the listener, while the speaker and the
listener were always positioned in the same places. The speaker and the listener were both
represented by a humanoid figure and consistently labeled as “speaker” and “listener”
respectively. The three dogs were consistently represented by different clip art dogs. In metadiscussion with the consultant, the speaker, listener, and dogs depicted in the diagrams were
consistently described as being positioned within a “room”. Each of the dogs appeared in the
same position in any diagram in which it was depicted. Dog 1 was always positioned near the
speaker, Dog 2 was always positioned approximately halfway between the speaker and the
listener, and Dog 3 was always positioned near the listener. However, each dog was not depicted
as being present in every discourse situation. The consultant was instructed to assume that any of
17
the three dogs not depicted in a given diagram was nevertheless a dog within the common
ground of the speaker and the listener.
Each dog may be considered proximal or distal based on its position within the diagram.
Dog 1 is considered proximal to the speaker because it is depicted as being within the speaker’s
peripersonal space, as described by Kemmerer (1999), and is distal from the listener because it is
not within the listener’s peripersonal space. Similarly, Dog 3 is considered proximal to the
listener and distal to the speaker because it is depicted as being within the listener’s, but not the
speaker’s, peripersonal space. Dog 2 is outside of the peripersonal space of both the speaker and
the listener and is considered distal to both.
The consultant was shown each diagram independently of the other diagrams. While
looking at each diagram, the consultant was asked to listen to four individual sentences in Tajik4.
After listening to each sentence, the consultant was instructed to suppose that the figure in the
diagram labeled “speaker” had uttered that sentence to the figure in the diagram labeled
“listener”. The consultant was then asked to indicate which of the dogs represented in the
diagram, if any, the speaker would be referring to when uttering that particular statement. When
viewing diagrams in which not all three dogs were depicted, the consultant was instructed to
suppose that the dog or dogs not shown in the diagram were still part of the speaker's and
listener's common ground, and that they may be considered as potential referents in any cases in
which the referent of a particular sentence can or must be an entity not present in the immediate
physical context of the discourse.
4
A fifth sentence, in dodzad (he barked), was introduced alongside the third diagram that was presented to the
consultant. This fifth sentence was presented to the consultant alongside the original four sentences and was then
included for all subsequent diagrams, but due to time constraints the fifth sentence was not tested with the first two
diagrams that were shown to the consultant.
18
Because the four sentences that accompanied the diagrams were constructed by me as a
non-speaker of Tajik, each sentence was produced for the consultant before introducing the
diagrams in order to confirm that it was grammatical. Because in and v͡ai also function as third
person singular pronouns, the data discussed here will only concern the results indicated by the
consultant for sentence 2 and sentence 3 in (4) below, in which in and va͡ɪ are being used as
adjectival demonstratives.
(4)
1.
sag
dodzad
dog barked
A dog barked.
2.
in
sag
dodzad
this
dog barked
This dog barked.
3.
va͡ɪ
sag
dodzad
that
dog barked
That dog barked.
4.
va͡ɪ
dodzad
that
barked
He barked.
The goal of these elicitations was to identify the speaker’s range of use for the
demonstratives in and va͡ɪ . The speaker had repeatedly referred to a constraint on the use of the
distal demonstrative va͡ ɪ in the meta-discussion during prior elicitation sessions, based on what
she described as speaker and listener “knowledge” or “familiarity” of the referent. This
elicitation technique was used in order to test the application of that constraint for potential
referents that were known to both the speaker and the listener as part of their common ground,
but which were absent from the physical discourse situation. The speaker had stated during
previous elicitation sessions that va͡ɪ could only be used for a referent if it was familiar to the
speaker and the listener, but had not previously been asked about referents that were not
19
physically present with the speaker and listener in the discourse situation. This technique was
also intended to test the speaker’s use of va͡ɪ against her use of the proximal demonstrative in for
potential referents at different physical distances from the speaker.
4.1
Diagrams A and B
In this section I will consider data from two diagrams, which show that the referent of in
‘this’ must be present in the discourse situation and that the referent of va͡ɪ ‘that’ may be absent
from the discourse situation only if it is part of the common ground. The consultant was shown
Diagram A, which depicts a speaker and a listener standing on opposite ends of a background
with two dogs present. One was labeled Dog 1 and was positioned next to the speaker. The
second was labeled Dog 3 and was positioned next to the listener. The consultant was instructed
to assume that another dog, Dog 2 was an existing dog that was absent from the discourse
situation but was part of the common ground between the speaker and the listener. The
consultant was asked which dog in Diagram A the referent of the demonstrative phrases in
sentences 2 and 3 was. Table 1 shows her responses.
DIAGRAM A
20
TABLE1
Sentence
2
3
Consultant Response
In
sag
dodzad
This
dog
barked
va͡ɪ
that
sag
dog
dodzad
barked
Dog 1
Dog 2 (absent), Dog 3
For Diagram A, the consultant responded that the referent of va͡ɪ in the sentence va͡ɪ sag
dodzad / that dog barked could be either Dog 3 (positioned distal to the speaker and proximal to
the listener) or Dog 2 (absent from the discourse situation). The consultant made the additional
qualifying remark that the sentence could refer to the absent dog only if it was “a dog that you
and the listener already know” or “the particular dog that you and the listener know”. In this
situation, Dog 1 (positioned proximal to the speaker and distal to the listener) is not a possible
referent of va͡ɪ sag / that dog.
Diagram B depicts a speaker and a listener with two dogs present. One is labeled Dog 1
and is positioned next to the speaker. The other is labeled Dog 2 and is positioned at the
approximate midpoint between the speaker and the listener, in the foreground. The consultant
was instructed to assume that Dog 3, absent from the discourse situation, was an existing dog and
part of the speaker’s and listener’s common ground. Table 2 shows the consultant’s responses for
which dog in Diagram B the referent of the demonstrative phrases in sentences 2 and 3 was.
21
DIAGRAM B
TABLE2
Sentence
2
3
Consultant Response
in
sag
dodzad
this
dog
barked
va͡ɪ
that
sag
dog
dodzad
barked
Dog 1
Dog 2 (absent), Dog
3
For Diagram B, the potential referents of the sentence va͡ɪ sag dodzad / that dog barked
include Dog 2 (distal to speaker, distal to listener) and Dog 3 (absent from the discourse
situation). Additionally, the common ground restrictions applied to the absentee dog in Diagram
A (which was Dog 2) applies here to Dog 3, the absentee dog in Diagram B. That is, in order for
va͡ɪ sag to refer to the absent Dog 3, Dog 3 must be part of the common ground. As with the
situation shown in Diagram A, these things are all also true of the equivalent English phrase that
dog. In both cases, an English speaker using the phrase that dog can be referring to either Dog 2
or Dog 3, provided that the referent, if absent, is part of the common ground; but the referent of
that dog is never Dog 1 (the dog positioned proximally to the speaker in both cases).
These data show that the use of va͡ɪ is limited by a non-spatial constraint. All of the
possible referents of va͡ɪ are positioned far from the speaker, and in both cases the referent of in
22
‘this’ is positioned near to the speaker. However, in addition to identifying the referent of va͡ɪ
within the situation shown in the diagram, when asked about these situations the consultant
explicitly stated that it is important for the referent of va͡ɪ , if absent, to be an entity that both the
speaker and the listener are familiar with, or part of their common ground.
This was also emphasized in several meta-statements made by the consultant, in which
she was specifically asked about the nature of the constraints on the use of va͡ɪ , and whether
there is a constraint based on physical distance. The consultant responded that the physical
distance of the referent from the speaker is important, but that there is additionally a constraint
based on common ground. When a referent is present in the discourse situation, it can be referred
to using va͡ɪ even if it was not previously part of the common ground between the speaker and
the listener. This is because the speaker can use an accompanying gesture to indicate which one
the intended referent is. The consultant was asked how a listener would identify the referent of
va͡ɪ when the potential referents are a distal referent that is present but not part of the common
ground, and a second referent, absent from the discourse situation, but part of the common
ground. The consultant responded that if the speaker does not identify the referent present in the
discourse situation in some gestural way, then the listener would assume that the correct referent
was something absent from the discourse situation.
The Tajik grammar descriptions by Rastorgueva (1963), Perry (2005), and Windfuhr &
Perry (2009) describe the constraints on the usage of both in and va͡ɪ as only spatial. Along with
the previous examples, these meta-statements highlight the importance of the common ground
criteria for determining which demonstrative a speaker will use (or which referent a listener will
choose in response to the speaker’s choice of demonstrative).
23
4.2
Diagram C
In the third example, the consultant was shown Diagram C, which again depicts a
situation including a speaker, a listener, and two dogs. This time Dogs 2 and 3 were present. Dog
2 was positioned at the approximate midpoint between the speaker and the listener, in the
foreground, and Dog 3 was positioned near the listener. The consultant was instructed to assume
that a third dog, Dog 1 (used in other diagrams) was an existing dog that was not present in the
discourse situation, but was part of the common ground between the speaker and the listener.
Table 3 lists the consultant’s answers as to which dog the referent of the demonstrative phrases
in sentences 2 and 3 is.
DIAGRAM C
TABLE3
Sentence
2
3
Consultant Response
In
sag
dodzad
this
dog
barked
va͡ɪ
that
sag
dog
dodzad
barked
24
Dog 2
Dog 3
In the situation described by Diagram C, the consultant identified Dog 3 as the only
potential referent of va͡ɪ sag (proximal to listener, distal from speaker), but not Dog 2 (distal
from both speaker and listener). The referent of in sag in Diagram C is the closest dog to the
speaker, which is Dog 2. This would not be the case if an English speaker were using the
equivalent phrase that dog in Diagram C. In English, both Dog 2 and Dog 3 would be potential
referents of that dog. There is no potential referent of the English phrase this dog depicted in
Diagram C. This is unexpected because the demonstrative reference systems of English and
Tajik have been categorized in the same way as both being speaker-anchored systems making a
two-way distance distinction Levinson (2004).
4.3
Diagram D
In Diagram D all three dogs are depicted. Dog 1 is positioned next to the speaker, Dog 2
is positioned at the approximate midpoint between the speaker and the listener, in the
foreground, and Dog 3 is positioned next to the listener. Table 4 lists the consultant’s responses
for which dog the referent of the demonstrative phrases in sentences 2 and 3 is.
DIAGRAM D
25
TABLE4
Sentence
2
3
Consultant Response
in
sag
dodzad
this
dog
barked
va͡ɪ
that
sag
dog
dodzad
barked
Dog 1
Dog 2, Dog 3
In the situation depicted in Diagram D, the only referent of in sag ‘this dog’ was Dog 1
which is proximal to the speaker and distal from the listener. Although Dog 2, which is distal
from both the speaker and the listener, was the referent of in sag in Diagram C, the consultant
did not consider Dog 2 a potential referent of in sag in Diagram D. Instead, Dog 2 is a referent of
va͡ɪ sag ‘that dog’, along with Dog 3 (distal from speaker, proximal to listener). These data
indicate that Dog 2 is the referent of in sag only if it is the physically closest potential referent, as
it is in Diagram C, However, when there is another potential referent present in the discourse
situation and physically closer to the speaker than Dog 2 (such as Dog 1 in Diagram D) then Dog
2 cannot be the referent of in sag.
4.4
Diagram E
In the fourth example, the consultant was presented with Diagram E, which depicts a
speaker and a listener standing at opposite ends of a background, with only Dog 2 present. Dog 2
is positioned at the approximate midpoint between the speaker and the listener. The consultant
was instructed to assume that two other dogs, Dog 1 and Dog 3 (used in other diagrams) were
existing dogs that were part of the speaker’s and listener’s common ground. Diagram E is shown
26
below and Table 5 lists the consultant’s responses for which dog the referent of each of the
demonstrative phrases in sentences 2 and 3 are.
DIAGRAM E
TABLE5
Sentence
2
3
Consultant Response
in
sag
dodzad
this
dog
barked
va͡ɪ
that
sag
dog
dodzad
barked
Dog 2
Dog 1, Dog 3
In Diagram E, the referent of in sag is Dog 2 (distal from both speaker and listener). The
phrase va͡ɪ sag refers to either Dog 1 or to Dog 3. But va͡ɪ sag cannot refer to Dog 2. In a
language such as English, Dog 2 would be referred to as that dog, but would not be referred to as
this dog because it is not proximal to the speaker. Unless the speaker adds a context to the
discourse in order to redefine the boundaries of the this-space, this dog would not have a referent
in Diagram E.
27
In the previous diagrams presented to the consultant, the referent of va͡ɪ sag could be a
dog that is present in the discourse context, but a dog is never referred to as in sag if there is
another dog positioned closer to the speaker. In Diagram E, the dog that is present in the
discourse situation is not positioned within the speaker's peripersonal space, but is nevertheless
referred to with the proximal demonstrative, while the distal demonstrative is used to refer to an
absent dog that is part of the common ground.
It is possible that the Tajik in is not a proximal demonstrative, but is instead a basic
demonstrative that serves an indicating function. Similarly, according to this analysis, the data
show that the distal Tajik demonstrative va͡ɪ is used in situations in which there is more than one
potential referent in the discourse and serves the function of adding additional information to the
meaning of the demonstratives. In the case of va͡ɪ , it adds elements of meaning that are physical
or spatial. According to the consultant, the referent of va͡ɪ is not positioned proximally to or
“next to” the speaker, but the data show that the reverse is not true for in. While the referent of
va͡ɪ must be distal from the speaker, the referent of in is not required to be proximal to the
speaker. Va͡ɪ also adds meaning based on the common ground status of the referent. This
common ground constraint introduced with va͡ɪ specifically applies to referents that are not
present in the discourse situation. This is because a potential referent of va͡ɪ that is present in the
discourse situation (which includes any potential referent of a demonstrative that is not the one
most proximal to the speaker) automatically becomes part of the common ground when it is
perceived mutually by both the speaker and the listener, and such a referent can easily be picked
out from other potential referents through the use of gesture.
28
5. Analysis of Tajik Demonstrative Usage
In this chapter I analyze the central findings of the data presented in Chapter Four. The
data show that demonstrative expressions in Tajik function differently from the classification
based on the criteria of anchor and number of distance-based spatial distinctions, described in
Chapter Three. According to these criteria, Tajik has a speaker-anchored system which makes a
two-way distinction. The problem with this classification system is that it assumes the spatial
distinction to be based straightforwardly on the referent’s distance from the speaker anchor.
However, the Tajik data elicited for this thesis show that a referent positioned distally – outside
of the speaker’s peripersonal space – may be referred to with either a proximal or a distal
demonstrative, depending on other circumstances in the discourse situation. This requires
explanation because it does not function according to the expectations set by the description of
the spatial distinction as based on distance from the speaker anchor. This description leads us to
expect that a referent that is distal from the speaker in Tajik should always be referred to with the
distal demonstrative.
My analysis contains three main claims. First, I argue that contrary to previous work on
demonstratives in Tajik (Rastorgueva 1963, Perry 2005, Windfuhr & Perry 2009), the Tajik
demonstrative system ought not to be classified simply as speaker-anchored. Recall from
Chapter Three that in a speaker-anchored system, the spatial distinction is made relative to the
speaker’s location. However, the Tajik data collected for this thesis conflict with this view.
Instead, the data support the claim that the Tajik proximal and distal demonstratives describe a
referent in terms of its distance from the speaker relative to other potential referents in the
29
discourse situation, rather than simply its position in terms of its distance from the speaker of the
utterance.
My second claim concerns the role of common ground as a factor in the speaker’s
selection of a demonstrative and the listener’s interpretation of it. As described in Chapter Three,
common ground includes the set of things, entities, knowledge, and beliefs shared by the speaker
and the listener in a discourse situation. While the Tajik demonstratives are consistently
described in the descriptive literature on Tajik as being proximal or distal in meaning, there are
constraints on the use of the distal demonstrative va͡ɪ in Tajik, based on whether or not the
referent of the demonstrative is part of the common ground. As emphasized in meta-statements
made by the consultant during the elicitation of data, in order to be a referent of va͡ɪ , an entity
must be part of the speaker’s and listener’s common ground.
My third claim is that the demonstrative system in Tajik consists of one basic
demonstrative, in, the function of which is purely indicative, and that other elements of meaning,
such as spatially proximal or distal, or the requirement that a referent be part of the common
ground between the speaker and the listener, are assigned to the basic demonstrative only in
contrast to those meanings being associated with the second demonstrative, va͡ɪ . This
understanding of two-way demonstrative reference is described by Enfield (2003), who notes
that the additional elements of meaning added by the use of additional demonstratives are most
often interpreted as having spatial meaning, specifically as indicating a referent according to its
distance from the speaker. However, in the case of va͡ɪ , the meaning contrast that the second
demonstrative adds to the system of reference is not directly based on the referent’s distance
from the speaker, but instead the contrast is related to the referent of in’s position relative to the
potential referents of va͡ɪ .
30
5.1
Distance-Base vs Exophoric
The function of demonstratives has often been described as fundamentally and
underlyingly based on distance from some feature of the speech situation (Diessel 2006, Evans &
Green 2006, Hanks 2009, Coventry et al 2014). This is the distance-based interpretation of
demonstrative reference. The exophoric understanding of demonstrative reference describes the
use of demonstratives as being simply to pick out or indicate a referent that is physically present
in the discourse situation. In this usage, the demonstrative is often accompanied by the use of a
gesture, such as pointing or gazing toward the intended referent. This exophoric usage is
described as the original, basic and fundamental use of demonstratives. All other uses of
demonstratives are described as extensions of this original and basic exophoric use. Additionally,
extended or non-exophoric uses of demonstratives are described by Evans & Green (2006) as
underlyingly the same as the exophoric use. Their claim is that while the exophoric use involves
indicating a referent that is physically proximal to or distal from the speaker in a discourse
situation, the non-exophoric uses do the same thing in an abstract sense. Extended, nonexophoric uses of demonstratives are metaphorically “pointing” to an abstract or physically
absent referent in the same way that a speaker would physically point to a referent exophorically.
In these cases the exophorically used demonstratives of a two-way system are described as
pointing to a referent specifically based on whether it is proximal or distal to a speaker, while in
the extended, non-exophoric use they are described as “pointing” to a referent based on some
other abstract sense of “distance” from the speaker or the discourse situation, based on factors
such as salience or importance (Leonard 1982, Kresin 1998) within the discourse, or ownership
(Hanks 2009). According to this definition of their use, demonstratives can be understood as
functioning to resolve the question of how far the referent is from the system’s anchor, which
31
will vary depending on the language: from the speaker in a speaker-anchored system, such as
that of Tajik or English, from the listener in an addressee-anchored system, or from both the
speaker and the listener in a speaker/addressee-anchored system, such as that of Japanese.
In the description of demonstrative reference provided by Enfield (2003), it is proposed
that the basic function of demonstratives is not to make a specifically distance-based reference to
an entity, as is commonly described. Instead, Enfield (2003) suggests that exophoric
demonstratives indicate a referent based on its location within the discourse situation relative to
other potential referents that are physically present in the situation. This is the location-based
interpretation of demonstrative reference. According to this understanding of demonstrative
reference, the exophoric use of demonstratives doesn’t function to answer the question, how far
away is the referent from the speaker?, as it would in a distance-based speaker-anchored system.
Instead the exophoric use of demonstratives in the location-based system functions to answer the
question, which of the potential referents is the intended referent? According to this view, the
use of in and va͡ɪ in Tajik is based on the distance of the referent from the speaker relative to the
location of other potential referents.
While in some examples of the data elicitation, it is consistent that the common
interpretation of the use of exophoric demonstratives as being distance-based can be applied to
Tajik, such an interpretation fails to account for the data that were collected regarding the
reference to Dog 2. Dog 2 was positioned outside of the peripersonal space of both the speaker
and the listener as a speaker- and addressee-distal potential referent. However, Dog 2 was not
always referred to using the distal demonstrative. Whether Dog 2 was a referent of the proximal
demonstrative or the distal demonstrative was dependent upon where other potential referents
were positioned within the discourse situation. However, the location-based interpretation
32
presented by Enfield (2003) accounts for the Tajik data collected for this thesis. Nevertheless, if
it is taken to be the correct interpretation for all languages, Enfield’s (2003) location-based
interpretation does not account for the differences in the speaker’s choice of demonstrative
between Tajik and English in some situations. In these cases, the use of demonstratives in
English is better described according to the distance-based interpretation.
These data imply that in Tajik the same referent can be identified as either "proximal" or
"distal" when it is the same distance from the speaker and the listener. For a Tajik speaker Dog 2
can be the referent of either in, as in Diagram C, or va͡ɪ , as in Diagram D. This would not be the
case in English, and Dog 2 would consistently be referred to using the same demonstrative, that,
in both Diagram C and Diagram D. This is because in both diagrams it is distal from the speaker
assuming that the deictic center is egocentric, meaning it is centered on the physical location of
the speaker (as is the case in English, and is most often the case in general, according to
Pederson et al 1998, Diessel 2006, and Hanks 2009).
The fact that the demonstrative used to refer to Dog 2 when positioned distally from both
the speaker and the listener differs between proximal in and distal va͡ɪ when neither the referent
nor the discourse participants change location suggests that distance is not the only factor and not
the most important factor in the selection of a demonstrative in Tajik. The classification of a
referent as what we would call “proximal” or “distal” for the purposes of talking about
demonstrative reference is more flexible in Tajik than it is in English, allowing the proximal
Tajik demonstrative in to refer to the closest potential referent even if that potential referent
would be considered distal under other circumstances in Tajik and even if that potential referent
would be considered distal in English. In other words, Tajik demonstratives do not pick out a
33
referent based solely or primarily on the referent’s distance from the speaker, but instead they
indicate a referent based on its position relative to other potential referents.
The Tajik proximal demonstrative in is not used exclusively for what would usually be
glossed by linguists as a "proximal" referent. In English the referent of this must be within a
certain range of the speaker. If it is outside of that range it will be referred to using that. Entities
located within that certain distance of the speaker will be considered proximal, while entities
outside of that space are distal. For English speakers this is a fixed space. The boundary of this
space does not have to be the same distance from the speaker in every discourse situation. The
size of the area is defined by the context of an utterance. For example, if an English speaker is
talking about this room then the space in which an entity will be considered proximal can extend
to the boundary of the room in question. In English, if there is no potential referent located
sufficiently near to the speaker as to be within the boundary of this space, then there is no
referent available for the demonstrative this. This is why there is no referent of this dog in
English in Diagram C. Unless the boundary of the speaker’s this-space is redefined in the context
of the discourse (such as by the speaker talking about this half of the room as opposed to that
half of the room, for example), the speaker’s this-space is assumed to be egocentric, including
the space occupied by the speaker and the speaker’s immediate vicinity. This consists of the
speaker’s peripersonal space (or near space). This describes the area physically within the
speaker’s reach, and can be extended if the speaker uses tools (such as a stick or a pointer) to
physically extend their reach.
However, the Tajik demonstrative in can be used outside of what would be considered
the this-space of an English speaker, as long as there is no other potential referent that is
physically closer than the intended referent. In Tajik it does not appear to matter whether the
34
referent is proximal to the speaker in an absolute sense, but instead whether (a) it is present in the
discourse context and (b) it is proximal to the speaker relative to other potential referents. If both
(a) and (b) are true, the proximal demonstrative is used.
Together, these data suggest that the distance-based interpretation and the location-based
interpretation are possible alternatives to each other and that these alternatives constitute an
additional dimension for classifying demonstrative reference systems. This additional dimension
allows us to capture the difference in behavior of Tajik and English demonstratives, where the
traditional two-dimensional classification system does not. Specifically, while English and Tajik
both have two-way, speaker-anchored systems, the data suggest that demonstrative reference in
English is distance-based, while in Tajik it is location-based.
In many discourse situations, the choice of demonstratives will be identical in a distancebased system and a location-based system. Thus, in the data elicited for this thesis, there are a
number of situations in which the consultant’s choice of proximal or distal demonstrative was
identical to what it would be for a speaker of English. The distance-based interpretation of
exophoric demonstratives was able to account for the answers given by the consultant while
viewing Diagrams A (p. 17), B (p. 19), and D (p. 23). This is because Diagrams A, B, and D all
depict a discourse situation in which a referent (Dog 1) is positioned proximally to the speaker.
i.e., within the speaker’s peripersonal space, and distally from the listener. Because the system of
demonstrative reference in Tajik is speaker-anchored, it is expected according to the distancebased interpretation that a referent of the proximal demonstrative in would be positioned
proximally to the speaker. And in accordance with this expectation, it is the case that in
Diagrams A, B, and D, the only entity that can be considered a potential referent of in is the
proximal dog: Dog 1. This would also be true of any variation of the discourse situation shown in
35
the diagrams that includes Dog 1. However, the distance-based interpretation does not account
for Diagram C (p. 22) or Diagram E (p. 27). It also fails to account for the difference in the
consultant’s answers to questions about what can potentially be the referent of the proximal in in
Diagram B and Diagram C. In Diagram B the referent of in is the proximal Dog 1, while in
Diagram C there is no proximal dog present and the referent of in is Dog 2, which is distal from
the speaker and the listener.
Neither Diagram C nor Diagram E depict the speaker-proximal Dog 1 in the discourse
situation. In a speaker-anchored system that functions according to the distance-based
interpretation, only a potential referent that is proximal to the speaker would be expected to be a
referent of the proximal demonstrative. This is because the listener’s identification of the referent
in such a system must be specifically based on the referent’s distance from the speaker. In such a
system, the proximal demonstrative would be used to indicate a referent that is proximal to the
speaker, within the speaker’s peripersonal space, and a referent that is distal from the speaker
would be indicated by using the distal demonstrative, regardless of where any of the other
potential referents are located. This should mean that in situations in which there is no speakerproximal potential referent, there would also be no available referent for the proximal
demonstrative. In such situations, all potential referents, being distal from the speaker, would be
expected to be referents of the distal demonstrative within a distance-based system. However, in
Diagram C, which shows two potential referents – one of them, Dog 2, which is distal from both
the speaker and the listener, and the other, Dog 3, which is distal from the speaker and proximal
to the listener – the referent of the Tajik proximal demonstrative in is Dog 2. Similarly, in
Diagram E, which shows only one potential referent – Dog 2, which is distal from both the
36
speaker and the listener – Dog 2 is a potential referent of the proximal in, but is not a potential
referent of the distal demonstrative va͡ɪ .
The distance-based interpretation by itself also fails to provide an explanation for the
differences in the use of demonstratives between Tajik and English. Both Tajik and English have
speaker-anchored systems of demonstrative reference. Both languages also make a two-way
distinction described in the literature as a proximal demonstrative (in in Tajik, this in English)
and a distal demonstrative (va͡ɪ in Tajik, that in English). Because of these similarities, one
might expect the use of demonstratives in Tajik and English to be identical. Indeed this view
seems to be generally accepted (Rastorgueva 1963, Perry 2005, Windfuhr & Perry 2009).
However, the data presented in Diagram C and Diagram E contradict the expectation that the
Tajik and English demonstratives have identical patterns of use.
Both of the potential referents in the discourse situation shown in Diagram C are distal
from the speaker. While in English both dogs would be referred to using the distal demonstrative
that, as would be expected according to the distance-based interpretation, the consultant
indicated that in Tajik only one of the dogs, Dog 3, would be referred to using the distal
demonstrative va͡ɪ . Dog 3 was shown in a position distal from the speaker and proximal to the
listener. The other dog shown in Diagram C, Dog 2, was positioned approximately halfway
between the speaker and the listener, outside of the peripersonal space of both. Despite not being
proximal to the speaker, the consultant indicated that Dog 2 was a potential referent of in but not
of va͡ɪ . Similarly, Diagram E shows only one potential referent: Dog 2, which was again
positioned approximately halfway between the speaker and the listener, distally from both. In
this situation in English, Dog 2 is a potential referent of distal that, but not a potential referent of
proximal this. In Tajik, the opposite is true: Dog 2 is a potential referent of proximal in, but not a
37
potential referent of distal va͡ɪ . If the distance-based interpretation of demonstrative reference
applied both to Tajik and to English, these differences in their use of demonstratives should not
exist.
On the other hand, the location-based interpretation of demonstrative reference is able to
account for all of the data collected in Tajik. According to the location-based interpretation, a
demonstrative indicates its referent, not simply based on how far away that referent is from the
speaker or listener as the anchor, but on its location relative to other potential referents.
In the case of Tajik, this means that the proximal demonstrative in doesn’t point to a
referent simply on the basis of being physically near to the speaker. Instead, if two or more
potential referents are present, in is used in Tajik to refer to whichever potential referent is
positioned nearest to (or least far away from) the speaker, without requiring that it actually be
within the speaker’s peripersonal space. This is why in Diagram C, in is used to refer to Dog 2
and va͡ɪ is used to refer to Dog 3. In Diagram C, Dog 2 and Dog 3 are both positioned distally
from the speaker, but Dog 3 is positioned more distally. When there is no proximal option, in
refers to the most proximal/least distal potential referent; in Diagram C, this is Dog 2. However,
when there is another potential referent that is positioned closer to the speaker than Dog 2, Dog 2
is no longer a potential referent of in, even though Dog 2’s physical distance from the speaker
has not changed. This is the case in Diagram D which depicts Dog 1 positioned proximally to the
speaker, Dog 2 positioned distally, and Dog 3 positioned more distally than Dog 2. In this
Diagram, in is used to refer to Dog 1 and va͡ɪ is used to refer to Dog 2 and Dog 3.
38
This is also why in Diagram E, Dog 2, the only dog present in the discourse situation, is referred
to using in. Despite not being positioned near to the speaker, Dog 2 is the referent of the
proximal demonstrative because there is no other potential referent that is positioned more
proximally to the speaker than Dog 2.
However, the location-based interpretation alone does not account for the differences
between Tajik and English in their selection of a demonstrative to refer to the only potential
referent present in Diagram E. This is why, while the location-based interpretation can be applied
to Tajik, it cannot be assumed to be universally applicable. If either interpretation were
universally applicable, demonstrative reference would be expected to share the same usage in
both Tajik and English, due to their similarities in terms of the anchor and the number of
distinctions. Instead, it appears to be the case that the distance-based interpretation applies to
English and that the location-based interpretation applies to Tajik. So, the application of the
location-based interpretation to Tajik does not show that the distance-based interpretation is a
wrong or false understanding of demonstrative reference; what these data show is that the
location-based interpretation and the distance-based interpretation are alternatives that exist
alongside each other and can be used to describe different languages.
5.2
Common Ground Constraint
Clark et al (1983) have described the importance of common ground in the interpretation
of demonstrative reference. Common ground refers to the collection of mutual knowledge,
experience, and beliefs shared by the participants of a discourse. Information can be part of the
common ground for a number of different reasons, including shared past experience, shared
39
perception, or mutual membership within a community or in-group. However, while it is possible
for information to be mistakenly identified by one discourse participant as being part of the
common ground, common ground information is generally known to all participants of the
discourse. It is possible for one discourse participant to mistakenly believe that some piece of
knowledge is part of the common ground, but information would not be considered part of the
common ground by any discourse participant unless they believed it to be mutually known
information between the discourse participants. Clark et al (1983) state that a listener refers to
the information in the common ground that they share with the speaker in order to identify the
intended referent of a demonstrative. The feature of common ground is not included in the
accounts of Tajik by Rastorgueva (1963), Perry (2005), or Windfuhr & Perry (2009). However, I
argue that the common ground constraint is an important element in understanding the use of the
Tajik distal demonstrative va͡ɪ , based on the consultant’s meta-statements during the data
elicitations.
As noted in Section 4.1 above, the consultant repeatedly made reference to a constraint
on the spatial use of va͡ɪ requiring that the referent of va͡ɪ must be part of the speaker’s and
listener’s common ground. This was described during the course of the data elicitation as a
requirement that the referent be “known” or “familiar” to both the speaker and the listener.
This constraint is most important when the referent of va͡ɪ is absent from the discourse
situation. This is because everything that is present in a discourse situation is by default part of
the common ground between the speaker and the listener by virtue of it being part of their shared
experience at the time of the discourse event. While a listener may refer to the information in the
common ground when identifying a referent that is present in the discourse situation, it is not
technically necessary for them to do so, since the speaker is also able to (and may be considered
40
likely to (Diessel 2006)) physically identify the referent by gesturing or gazing toward it. In other
words, the speaker points to the literal common ground to identify a figure. However, when the
referent of va͡ɪ is absent from the discourse situation, the listener must rely on information in the
common ground in order to identify the most likely referent (Clark et al 1983).
5.3
Basic Demonstrative, Meaning Contrast, and the Dual Function of va͡ɪ
Enfield (2003) presents the view that, rather than each term within a two-way system of
demonstrative reference having an inherent meaning that contrasts with the other, every language
has one basic demonstrative which functions solely indicatively, and that the contrasting
meanings that are carried by the complementary demonstratives in a language are the result of
additional demonstratives. These additional demonstratives, Enfield argues, are added in order
to create a way to distinguish between more than one potential referent. According to this view,
each language has a demonstrative that serves only the basic function of indicating a referent,
and all other meanings associated with that demonstrative, such as proximal or distal, are due to
the introduction of additional terms, which have contrasting meaning.
This view can be applied to demonstrative reference in Tajik in a way that is supported
by the previous two claims made in this section: that Tajik uses a location-based system of
demonstrative reference, and that the use of the distal demonstrative va͡ɪ in Tajik requires the
referent to be part of the speaker’s and listener’s common ground. The same evidence, regarding
the use of the proximal in, that demonstrates a location-based system in Tajik also supports the
idea that in is the basic or unmarked demonstrative in Tajik because this evidence shows its
distribution to be wider and more versatile than that of the distal va͡ɪ . Additionally, the fact that
the common ground constraint only applies to va͡ɪ is also compatible with this view because it
41
demonstrates the addition of a meaning contrast alongside the inclusion of va͡ɪ in the
demonstrative reference system.
The location-based system of reference in Tajik was supported by several pieces of data
which are related to the speaker’s selection of a demonstrative in a case when the intended
referent is present in the discourse situation, but not proximal to the speaker, and when there is
no other potential referent positioned between the intended referent and the speaker. In this case,
the intended referent, while not actually being positioned near the speaker, can be considered the
“most proximal” potential referent available. In every case, the data show not only that a
potential referent that is proximal to the speaker would be referred to using in, but also that in the
absence of a proximal potential referent within the speaker’s peripersonal space, the most
proximal referent available would be referred to using in. This is also the case when there is only
one potential referent available in the discourse situation, as in Diagram E (p. 27). If there was
only one dog pictured in the diagram, the consultant indicated that it was the referent of in but
was not a referent of va͡ɪ . This shows that in the most basic situation, when there is no need to
differentiate between one potential referent and another, the demonstrative used to indicate a
referent is always in, regardless of the referent’s actual proximity to the speaker.
There are two ways in which the basic situation can be changed in order to create a
circumstance under which the intended referent needs to be distinguished from some other
potential referent(s) and the other type of Tajik demonstrative (va͡ɪ ) will be used. The first way
is to add additional potential referents to the physical discourse situation. This was the case in
most of the diagrams presented to the consultant, that is, most included more than one dog. The
other way is to include additional potential referents that are not present in the discourse
situation. These are the only two circumstances in which there is a potential referent for va͡ɪ .
42
There are two meanings associated with va͡ɪ which make it a contrasting term of in. Both
meanings were explicitly identified by the consultant during the elicitation sessions. The first
meaning of va͡ɪ is spatial in nature, and it is a distal meaning. In the meta-linguistic discussion,
the consultant described the use of va͡ɪ as limited to referring to an entity that is “not next to [the
speaker]” or “away from [the speaker]”. Va͡ɪ is then correctly glossed as a distal demonstrative.
However, in is not a proximal demonstrative in this case; it simply refers to a proximal referent
in situations when it is used to distinguish between a referent of va͡ɪ and a proximal entity.
The other meaning associated with va͡ɪ has to do with the common ground. The
consultant stated that the referent of va͡ɪ must be part of the common ground, or must be
“something in common that [the speaker] and the listener have or know”. This common ground
constraint becomes important for the speaker to consider when the referent of va͡ɪ is absent from
the discourse situation. This is because any potential referent that is present in the discourse
situation is automatically in the speaker’s and listener’s common ground. Because they are in a
situation with the potential referent, it becomes part of their mutually shared experience, and in
that case va͡ɪ can be interpreted as simply having the spatial function of being distal from the
speaker. It is only when the intended referent of va͡ɪ is absent from the discourse situation that it
becomes necessary for the referent to have already been established within the speaker’s and
listener’s common ground.
Both of these meanings only apply to va͡ɪ . Only va͡ɪ specifies that its referent is distal, in
the case of a referent present in the discourse situation, and only va͡ɪ specifies that its referent is
part of the common ground of the discourse participants, in the case of a referent absent from the
discourse situation. When a speaker uses in in a situation such as that in Diagram E, when there
is only one potential referent, in’s function is simply to indicate or “point to” that potential
43
referent. When a speaker uses in in any other situation, its function is not to specify a referent
that is proximal, or even a referent that is most proximal. Instead the function of in is to specify
that its referent is the potential referent that is not va͡ɪ . The meanings of va͡ɪ are “distal” or
“absent but part of the common ground”, and the meaning of in is “not va͡ɪ ”, which can be
specified as “not the one that is distal” and “not absent but part of the common ground”.
In this chapter, I have interpreted the data presented in the previous chapter as implying
that (1) Tajik uses a location-based system of demonstrative reference, not a distance-based
system; (2) the demonstrative glossed as “distal” in Tajik, va͡ɪ , does not refer only to a referent
that is physically distal from the speaker, outside of their peripersonal space, but also requires its
referent to be part of the common ground between the discourse participants; and (3) the
location-based meaning and the common ground constraint only apply to va͡ɪ , while the function
of in is simply to indicate a referent. This interpretation allows us to provide an analysis that
accounts for all of the data elicited from the consultant for this project. Interpreting the data as I
have done also provides an explanation for the differences between English and Tajik choices of
demonstratives in identical discourse situations, despite their both being two-way speakeranchored systems.
44
6. Conclusions, Implications, and Issues for Further Research
The principle claim of this thesis is that the description of demonstrative reference in the
existing literature on Tajik does not adequately describe the use of demonstratives in Tajik. The
systems of demonstrative reference in Tajik and English are very similar, and this is why the
existing descriptive literature on Tajik describes demonstrative reference in such a way that it
seems to work exactly the same way in both languages. In the descriptive literature on Tajik, the
two types of demonstratives are described as proximal (in) and distal (on and va͡ɪ ) and are
glossed as the English counterparts ‘this’ and ‘that’ respectively (Rastorgueva 1963, Perry 2005,
Windfuhr & Perry 2009). Without additional, specific, or more detailed information about the
use of demonstratives provided in the descriptive literature on Tajik, demonstrative reference in
Tajik appears to be done in exactly the same way that it is in English, in which the speaker’s
choice of demonstrative is based on the referent’s distance from the speaker. However, the data
presented in this thesis show a significant difference between the selection of a demonstrative in
Tajik and in English for the same referent in some identical situations. The common factor
shared between these situations in which Tajik and English are contradictory is the lack of a
potential referent that is proximal to the speaker. In English, when all of the potential referents
are distal from the speaker, the speaker will use the distal demonstrative that to refer to any
potential referent. In Tajik, on the other hand, when all of the potential referents are distal from
the speaker, the speaker will use the proximal demonstrative in for whichever potential referent
is least distal, and the distal demonstrative va͡ɪ for any other potential referents that are more
distal than the referent of in.
45
6.1
Categorizing Systems of Demonstrative Reference
The common general descriptions of demonstrative reference provided in the literature
are also insufficient to differentiate between the demonstrative reference systems of Tajik and
English. This can potentially cause confusion about the differences between Tajik and English.
In addition, using the overly broad categorization found in much of the existing literature on
Tajik could further cause potential confusion about the differences in demonstrative reference
between any two languages. This could lead to incorrect conclusions being drawn about Tajik,
based solely on information in the available literature, and incorrect conclusions being drawn
about demonstrative reference in general, based on the commonly used classification criteria
being applied to insufficient information about the use of demonstrative reference in any
particular language.
Based on the commonly described features of demonstrative reference in terms of anchor
(speaker, addressee, or both) and the number of distance-based spatial distinctions made (two,
three, or more) (Levinson 2004, O’Grady 2010) Tajik and English may be expected to be
grammatically identical in terms of their use of demonstratives. Described according to those two
features, both languages have a speaker-anchored system that makes a two-way distance
distinction. Because the spatial distinction is based on the referent’s distance from the speaker,
this system does not account for the differences between Tajik and English discussed above.
According to a distance-based classification, a referent that is distal from the speaker should be
referred to using the distal demonstrative whether it is the most or least distal potential referent.
This is the case in English. However, the data show that it is not the case in Tajik. Applying the
location-based interpretation of demonstrative reference as described by Enfield (2003) to Tajik
accounts for this by interpreting demonstrative reference in a way that is not based on the
46
referent’s distance from the speaker, but instead on the referent’s location relative to other
potential referents. Instead of using the distal demonstrative for all distal referents because of
how far away those referents are is from the speaker, a Tajik speaker uses the proximal
demonstrative for the least distal potential referent because of where the other potential referents
are located. Similarly, a Tajik speaker uses the distal demonstrative for all potential referents that
are more distal than the least distal one.
However, the location-based interpretation provided by Enfield (2003) does not account
for demonstrative reference in every language. A location-based understanding of demonstrative
reference explains the behavior of demonstratives in Tajik, but is not able to account for the
behavior of demonstratives in English. Instead, the use of demonstratives in English can be
accounted for by the more commonly described distance-based interpretation. This means that
neither interpretation by itself can account for demonstrative reference in all languages, but that
an individual language may be categorized as either distance-based or location-based. Rather
than assuming that only one of the two interpretations -- a distance-based system or a locationbased system – is correct, allowing for both types as possible alternatives to each other accounts
for these differences between Tajik and English.
6.2
The Role of Common Ground in Demonstrative Reference
The specific details of the use of va͡ɪ in Tajik require certain common ground constraints
to be met by its referent. In the existing literature on Tajik, va͡ɪ is glossed as the English that and
is described as a distal demonstrative. However, simply stating that va͡ɪ is used for a referent that
is positioned far away from the speaker does not accurately describe the use of va͡ɪ by the Tajik
speaker. According to the data I collected, this is an inaccurate characterization of va͡ɪ ’s use.
47
This is not only because of the previously mentioned differences between Tajik and English due
to the distance-based vs. location-based distinction. It is also because va͡ɪ carries an additional
meaning that is unrelated to the physical position of its referent. In order to be a referent of va͡ɪ ,
the entity in question is required to be part of the common ground. This is easily resolved for
referents that are present in the discourse situation, and could be easily overlooked in such
situations because any potential referent that is present in the discourse situation is automatically
a part of the speaker’s and listener’s common ground. If every potential referent is present, then
every potential referent is part of the common ground and the common ground constraint on the
use of va͡ɪ is unimportant. In such situations the only distinction that needs to be made to
differentiate between potential referents is the location-based spatial distinction. However, in
situations that include a referent that is absent from the discourse situation, the common ground
constraint becomes important. An absent referent can be referred to using va͡ɪ if it is already part
of the common ground, or if it is immediately introduced to the common ground when it is
referenced. This is not unique to Tajik - it is also the case in English - but it is not mentioned in
any of the literature on Tajik.
6.3
In as a Basic Demonstrative
The conclusions drawn in the previous two sections also support the idea that Tajik
behaves according to the way Enfield (2003) describes demonstrative reference systems, as
consisting of a basic or default demonstrative that has a purely indicative function, as well as a
number of additional demonstratives (depending on the number of distance- or location-based
distinctions made by the language in question). These additional demonstratives carry additional
48
meanings adding a meaning contrast to the system in situations when the additional
demonstratives are used. According to this view, the contrasting distance-based meaning is
carried by the non-basic additional demonstrative. The second, non-default demonstrative in
Tajik is va͡ɪ . Because the data show the common ground constraint applying only to va͡ɪ , and
because the consultant repeatedly made specific mention of this constraint’s application to va͡ɪ
but never mentioned this constraint with regard to the other demonstrative, in, it seems to be the
case that the common ground constraint only applies to va͡ɪ . This is because, while in can only
be used to point to a referent that is present in the discourse situation (and is therefore
automatically part of the common ground), va͡ɪ can be used for a referent that is absent. The
specific requirement that the referent be part of the common ground is part of the meaning of
va͡ɪ , which can be used to point to absentee referents only if they are part of the common ground.
But such a requirement does not apply to in, nor does it contribute meaning to in, which only
functions as an indicative word which simply points to a referent.
6.4
Implications
Because Tajik is an understudied language, the available literature on Tajik is minimal. In
the most general sense, this work adds to the small existing body of work written about Tajik in
English. Additionally, this work contributes a detailed description of the use of demonstratives in
Tajik, which is not present in any of the existing descriptive literature on the language. In
contributing a detailed description and proposing an analysis of demonstrative reference in Tajik,
this work also calls attention to potential shortcomings of available descriptions of demonstrative
reference in general. It additionally highlights the need for more research to be performed
regarding the systems of demonstrative reference with data from specific languages, and the
ways in which languages can be categorized according to their demonstrative reference systems.
49
While there are existing publications that discuss and explain demonstrative reference,
the ideas presented on demonstrative reference are generally not applied to language description.
This can lead to confusion or incorrect use of language in pedagogical situations and to incorrect
claims when basing academic work on information obtained from a reference grammar. Using
existing published descriptions of Tajik, there would be no way to reach the same conclusions
shown by the data in this thesis, and one would instead be likely to assume that Tajik
demonstratives are distance-based, as English demonstratives are. Similarly, while the concept of
common ground is present within the background literature on demonstrative reference in
general, it is not addressed in the literature on Tajik. The common ground constraint as it applies
to Tajik may or may not apply to other languages as well.
6.5
Limitations and Areas for Further Research
The circumstances of this research included a number of limiting factors. As a result, the
evidence represented in this work consists of a relatively small number of examples. In order to
strengthen the conclusions drawn in this work, a number of specific factors should be considered,
which would need to be addressed in further research on this topic.
This project took place over a relatively short span of time with a relatively small number
of elicitation sessions, with only a single language consultant. The data used for this work were
collected over a period of eight weeks as part of a course on linguistic field methods. During that
period of time, there were seven joint elicitation sessions conducted (either with a small group of
fellow students or during class time, observed by the entire class) that were relevant to this
50
project, and approximately twenty minutes of each of the seven hour-long elicitation sessions
were specifically devoted to collecting the data for this project.
In order to solidify the conclusions drawn by this work, further research would need to be
performed with a wider variety of Tajik speakers. Ideally, such a group would include speakers
with various characteristics, most importantly including both male and female participants, and
participants of various age groups.
Additionally, the original goal of this research was not specifically about the use of
demonstrative reference. Because of this, data items related to the original topic included very
few instances of demonstratives being used by the consultant. By the time the data showed
promise for revealing facts about demonstrative reference, there was very little time left to work
with the consultant. It was only for the last two data elicitation sessions that the data elicited for
this project consisted entirely of items relevant to demonstrative reference. Even following a
similarly limited time frame, conducting a similar research project while devoting the data
elicitation session time purely to the topic of demonstrative reference might yield more robust
data and lead to stronger conclusions.
Conducting the data elicitation required careful selection of the diagrams to be shown
during the elicitations. This is why only a very limited set of five separate diagrams were shown
to the consultant. In order to make the best use of the time available to further investigate
demonstrative reference, the situations shown in the diagrams presented to the consultant were
constructed with the intention of eliciting the most useful or interesting results, while avoiding
elicitation of redundant information. When performing further research on this topic, it would be
worthwhile to include a similar set-up involving diagrams depicting various discourse situations,
51
including all potential configurations of all three potential referents in the speaker-proximal,
medial, and listener-proximal positions used for the original research.
In addition, including diagrams involving more and different potential referents could
also yield useful results. For example, one could perform similar data elicitation research using
potential referents that are of the same type or category as the others (such as all dogs), but
which differ significantly in terms of physical size. Additionally, one could perform this type of
research while adding details to the discourse situation, such as a table or a doorway, that might
act as a boundary of some sort between two potential referents.
Testing similar situations on a smaller scale could also strengthen the conclusions drawn
by this work. This might include using some similar configurations of small items as potential
referents (one speaker-proximal, one approximately at the midpoint between a speaker and a
listener, and one listener-proximal) on a table with a speaker and a listener on either side of the
display of objects. Similarly, one might then test the same potential referents similarly placed in
various configurations on a table for a speaker and listener who are both positioned near each
other, such as on the same side of or along the same edge of the table.
This work also did not include any questions about the third Tajik demonstrative, on, and
did not address it in any way. On is described by Rastorgueva (1963), Perry (2005), and
Windfuhr & Perry (2009) as a distal demonstrative, the use of which is identical to that of va͡ɪ .
According to the background literature on Tajik, on and va͡ɪ can be used interchangeably. The
main reason for the exclusion of on in this project was because the responses elicited from the
consultant for the entire duration of the eight weeks over which the elicitation sessions took
52
place never included any usage of on. Further research might test whether on can be substituted
for va͡ɪ in the situations where the consultant for the present research used va͡ɪ .
Doing the same type of project in other languages would also provide additional data
about demonstrative reference in general. According to the data collected for this project, the
categorization and general behavior of demonstrative reference in Tajik appear to be very
different in their actual use than what could be predicted based on the background literature.
There is no existing description of Tajik grammar written in English that sufficiently explains its
system of demonstrative reference and no data examples were provided that would lead a reader
to question its categorization as distance-based.
It is possible that this is also the case in other languages. Because demonstrative
reference is rarely addressed specifically in any great amount of detail in grammatical
descriptions, it is possible that demonstratives as described in a given language are assigned
distance-based meaning in the absence of sufficient data and examples to support the distancebased interpretation, or in absence of sufficient data and examples to specifically rule out the
location-based interpretation for that language. In many cases, demonstrative reference is not
even mentioned in that amount of detail, and the distance-based interpretation is not well
explained, if at all; demonstrative reference is instead described in grammatical descriptions
simply by glossing the demonstratives as their nearest and most accurate English equivalent.
Glossing demonstratives makes sense for the purposes of translation, but doing so in
absence of a full explanation of the system of demonstrative reference in a given language can
lead to confusion on the part of the reader, such as misuse or ungrammatical use of the language
(in pedagogical situations), or drawing erroneous conclusions about the use of demonstratives or
categorization of demonstrative reference based on the information presented in a grammatical
53
description. This potential for error and confusion supports the validity and necessity of
performing the type of research presented in this paper, in order to minimize the possibility of
such situations.
Finally, the current thesis is focused on data elicited from a consultant that was
exclusively about spatial demonstrative reference. In all of the discourse situation diagrams
presented to the consultant, the potential referents were depicted as physically tangible entities
that were present in the discourse situation with the speaker and the listener. However,
demonstrative reference is not exclusively spatial or limited to referents that are physically
present in the discourse situations, as demonstrated by the Tajik data relating to the common
ground constraint for reference to entities absent from the discourse situation. Further research
related to the use of demonstrative reference for abstract or intangible entities, in Tajik or in any
language, could further support conclusions about the categorization of that language’s
demonstrative reference system, or could lead to alternative analyses about the use of
demonstratives within a language.
54
References
Clark, H. H., Schreuder, R., & Buttrick, S. (1983). Common ground and the understanding of
demonstrative reference. Journal of Verbal Learning and Verbal Behavior, 22, 245 - 258.
Coventry, K. R., Griffiths, D., & Hamilton C. J. (2014). Spatial demonstratives and perceptual
space: describing and remembering object location. Cognitive Psychology, 69, 46 – 70.
Diessel, H. (2006). Demonstratives, joint attention, and the emergence of grammar. Cognitive
Linguistics, 17 (4), 463-489.
Enfield, N. J. (2003). Demonstratives in space an interaction: data from Lao speakers and
implications for semantic analysis. Language, 79(1), 82 – 117.
Evans, V. and Green, M. (2006). Cognitive Linguistics: An introduction. Mahwah, NJ: Lawrence
Erlbaum Associates, Inc.
Hanks, W. F. (1992). The indexical ground of deictic reference. In A. Duranti & C. Goodwin
(Eds.) Rethinking Context: Language as an interactive phenomenon (43 – 76). New
York: Cambridge University Press.
Hanks, W. F. (2009). Fieldwork on deixis. Journal of Pragmatics, 41, 10 – 24.
Kemmerer, D. (1999). “Near” and “far” in language and perception. Cognition, 73, 35 – 63.
Kresin, S. C. (1998). Deixis and thematic hierarchies in Russian narrative discourse. Journal of
Pragmatics, 30, 421 - 435.
Leonard, R. A. (1982). The Semantic System of Deixis in Standard Swahili (Doctoral
Dissertation).
Levinson, S. C. (2004). Deixis. In L. R. Horn & G. Ward (eds.) The Handbook of Pragmatics (97
– 121).
Maieborn, C., von Heusinger, K., & Portner, P. (2012). Semantics. Berlin: Mouton de Gruyter.
O’Grady, W. (2010). Semantics: the analysis of meaning. In O’Grady, W., Archibald, J.,
Aronoff, M., & Rees-Miller, J. (eds.) Contemporary Linguistics (203 – 244).
Pederson, E., Danziger, E., Wilkins, D., Levinson, S., Kita, S., & Senft, G. (1998). Language,
74(3), 557 – 589.
Perry, J. R. (2005). A Tajik Persian Reference Grammar. Boston, MA: Brill.
Rastorgueva, V. S. (1963). A Short Sketch of Tajik Grammar. (H. H. Paper, Trans. & Ed.). The
Hague: Mouton & Co.
Tajiki. (n.d.) In Ethnologue. Retrieved from https://www.ethnologue.com/language/tgk
Windfuhr, G. & Perry, J. R. (2009). Persian and Tajik. In W. Gernot’s (Ed.) The Iranian
Languages. (416-544).
55